**Caseinate-Stabilized Emulsions of Black Cumin and Tamanu Oils: Preparation, Characterization and Antibacterial Activity**

#### **Lucie Urbánková 1,\*, V ˇera Kašpárková 1,2, Pavlína Egner 1, Ondˇrej Rudolf <sup>1</sup> and Eva Korábková <sup>2</sup>**


Received: 3 November 2019; Accepted: 25 November 2019; Published: 27 November 2019

**Abstract:** Caseinate-stabilized emulsions of black cumin (*Nigella sativa*) and tamanu (*Calophyllum inophyllum*) oils were studied in terms of preparation, characterization, and antibacterial properties. The oils were described while using their basic characteristics, including fatty acid composition and scavenging activity. The oil-in-water (o/w) emulsions containing the studied oils were formulated, and the influence of protein stabilizer (sodium caseinate (CAS), 1–12 wt %), oil contents (5–30 wt %), and emulsification methods (high-shear homogenization vs. sonication) on the emulsion properties were investigated. It was observed that, under both preparation methods, emulsions of small, initial droplet sizes were predominantly formed with CAS content that was higher than 7.5 wt %. Sonication was a more efficient emulsification procedure and was afforded emulsions with smaller droplet size throughout the entire used concentration ranges of oils and CAS when compared to high-shear homogenization. At native pH of ~ 6.5, all of the emulsions exhibited negative zeta potential that originated from the presence of caseinate. The antibacterial activities of both oils and their emulsions were investigated with respect to the growth suppression of common spoilage bacteria while using the disk diffusion method. The oils and selected emulsions were proven to act against gram positive strains, mainly against *Staphylococcus aureus* (*S. aureus*) and *Bacillus cereus* (*B. cereus*); regrettably, the gram negative species were fully resistant against their action.

**Keywords:** black cumin (*Nigella sativa*) seed oil; tamanu (*Calophyllum inophyllum*) seed oil; emulsion; formulation; antibacterial activity

#### **1. Introduction**

In recent years, there has been increasing demand for the use of biopolymers as emulsion stabilizers in the cosmetic, pharmaceutic, and food industry. Additionally, alternative sources of bioactive substances, such as essential fatty acids, are in focus. Many of the natural products that were extracted from plants demonstrate biological activities, and they receive particular attention as a source of valuable fatty acids, antimicrobials, antioxidants, or wound healing agents [1,2].

*Nigella sativa*, which is commonly called black cumin (BC), and *Calophyllum inophyllum* (tamanu, TA), have been used for its nutritional and therapeutic value for many years. Black cumin and its seeds with a characteristic strong taste have been exploited for both culinary and medicinal purposes. The reason for the biological activity of cumin seeds (antioxidant, anticancer, anti-inflammatory, as well as antibacterial) can be found in their complex composition, as they contain over 100 different constituents, including all essential fatty acids. In general, BC consists of oil, proteins,

and carbohydrates [2,3]; oil from black cumin is rich in linoleic and oleic acids, as well as bioactive phytosterols and tocopherols [4]. In addition, the seeds contain active phytochemicals, such as thymoquinone, thymohydroquinone, q-cymene, carvacrol, and 4-terpineol [5]. This oil can be used as a natural resource for the production of pharmaceuticals and novel functional foods thanks to its composition and effects [2,6–8].

Tamanu oil expelled from the seeds of *Calophyllum inophyllum* is mainly beneficial in the treatment of dermal problems. The antibiotic and anti-inflammatory properties make this oil an excellent raw material for regenerating and protective formulations. Calophyllic acid and a lactone with antibiotic properties are the two main bioactive substances in the tamanu oil [9]. *Calophyllum inophyllum* is also known as a rich source of secondary metabolites (coumarins, xanthones, flavonoids, and triterpenes) [10], with some of them having anti-inflammatory, antibacterial, and antiviral properties [11,12]. Additionally, coumarin derivatives (calaustralin, calophyllolide, inophyllum, inophyllum E) that were obtained from a crude extract of the nuts were reported to have activity against *Staphylococcus aureus* [1] and antioxidant properties [13].

Nevertheless, the presence of unsaturated fatty acids in both oils causes their sensitivity to oxidation. This makes encapsulation a suitable strategy for protecting the oils, thus allowing for the delivery of lipophilic compounds into aqueous-based systems. For the efficient encapsulation, the production of stable emulsions is crucial and the properties of stabilizing surface layer, oil, as well as the emulsion characteristics are the central factors that can affect the performance of final product; emulsification also plays a key role in optimizing the encapsulation efficiency of oils [14,15].

Emulsification of triacylglycerol-based oils (fish, olive, or sunflower oil) is frequently conducted with milk proteins [16–20]. Particularly, sodium caseinate (CAS) is a common emulsifier and it can be the integral part of formulations due to its amphiphilic structure and surface activity. In o/w emulsions, CAS has been used as emulsifier and encapsulating agent [21], as it forms a barrier by adsorbing to the oil-water interface. This barrier is essential for protecting the bioactive substances against oxidation and the molecules that are adsorbed at oil-water interface also provide an effective shield against flocculation and coalescence due to a combination of electrostatic and steric repulsion. For example, CAS provided good protection of fish oil against oxidation by forming a physical barrier effect [16,22–24]. Casein is also able to protect other oils against oxidation. As published by Hu, et al., corn o/w emulsions that were stabilized with CAS showed both high physical and oxidative stability, which was probably due to ability of CAS to produce a thick layer on emulsion droplet interface and its unique chelating properties [25].

The above summary clearly shows that the incorporation of rarely studied tamanu and black cumin oils in emulsions can be of benefit, as they contain active ingredients with biological and pharmaceutical effects (treatment of asthma, bronchitis, skin diseases; antimicrobial, anti-inflammatory effects; and, gastro-protective properties) [8,9,26,27]. Moreover, the stabilizing of emulsions containing these oils with biocompatible and biodegradable protein CAS can facilitate the delivery of oils to hydrophilic systems while also protecting them against degradation. Therefore, the aim of the work was to prepare o/w emulsions of tamanu and black cumin oil stabilized with CAS, which can serve as carrier systems with therapeutic and physiological benefits for humans. The study also aimed at finding the optimum emulsion formulation with respect to the concentrations of both CAS and oils, and at establishing a suitable emulsification procedure when using two commonly available methods, high-shear homogenization and sonication. The novelty of the study lies in the determination of antimicrobial properties of the oils and their emulsions against both gram positive and gram negative strains, as contamination with spoilage microorganisms is one of the crucial problems encountered in cosmetic and food industry. In the view of the fact that the studies that deal with the emulsification of black cumin and tamanu oils are relatively scarce and only limited data on this topic is found in the scientific literature, the presented study is original and it brings about topical information in this area.

#### **2. Materials and Methods**

#### *2.1. Materials*

Casein sodium salt (CAS), DPPH (1,1-diphenyl-2-picrylhydrazyl), toluene, and sodium thiosulfate were provided from Sigma-Aldrich (Darmstadt, Germany). Non-traditional vegetable oils from *Calophyllum inophyllum* (Tamanu oil, TA) and *Nigella sativa* (Black cumin oil, BC) were obtained from Nobilis Tilia (Krásná Lípa, Czech Republic). All of the chemicals and reagents were of analytical grade and used without further purification.

#### *2.2. Characterization of Oils*

#### 2.2.1. Basic Characteristics

Iodine value (IV; Hanus) expressing the amount of unsaturation in fatty acids present in the oil, saponification value (SV), which is a measure of average molecular mass of all fatty acids, acid value (AV) corresponding to amount of free fatty acid in the oil, and peroxide value (oxidation stability) were determined while using volumetric analysis. The standard methods that were performed with slight modifications were followed [28].

#### 2.2.2. Fatty Acid Composition by Gas Chromatography-Flame Ionization Detector (GC-FID)

The composition of fatty acids in oils was determined by GC-FID after their conversion to respective methyl esters (FAME). In the case of tamanu oil, 30 mL methanol, 2 mL methanolic phenolphthalein solution, and 5 mL hexane were used to dissolve 2 g oil. The solution was heated for approx. 15 min. and the hot sample was then titrated with 0.5M methanolic KOH to neutralize the sample. The sample was added 1 mL 1M methanolic potassium hydroxide and then heated again for 30 min. After the reaction was completed, the FAMEs were twice extracted with 2 × 10 mL hexane. FAME of black cumin oil were prepared via mixing of 2 g oil with 20 mL methanol and 0.5 mL methanolic KOH (1 M). The solution was heated for 30 min. and the resulting products were twice extracted with 2 × 10 mL hexane. The samples of methyl esters were diluted with hexane for analysis.

GC analyses were conducted while using a DANI Master GC Fast Gas Chromatograph (DANI Instruments, Cologno Monzese, Italy) that was equipped with a flame ionization detector (FID) (DANI Instruments, Cologno Monzese, Italy) and capillary column Phenomenex ZebronTM ZB-5MS (30 m × 0.25 mm × 0.5 μm, Phenomenex Inc., Torrance, CA, USA). The injection volume was of 1 μL and flow of a nitrogen carrier gas of 1 mL·min<sup>−</sup>1. The temperature gradient started at 110 ◦C, followed by increase of 5 ◦C·min−<sup>1</sup> to 240 ◦C and then to 280 ◦C at the rate of 2 ◦C·min<sup>−</sup>1. The sample was held at the temperature for 20 min. The temperatures of injector and detector were set to 230 and 300 ◦C, respectively. The FAMEs were identified by comparing their retention times with those of *SUPELCOTM 37 Component FAME Mix* standard (Sigma Aldrich, Darmstadt, Germany). The quantification of FAME was performed according to internal normalization method. The results are reported in percentage (wt %) of respective fatty acid/100 g total fatty acid content.

#### 2.2.3. Antioxidant Activity

Free radical scavenging activity of BC and TA oils was assayed according to [4]. The DPPH radicals were dissolved in toluene at concentration of 10−<sup>4</sup> M. For evaluation, 10 mg of oil (in 100 μL toluene) was mixed with 390 μL toluenic DPPH• solution and then vortexed for 20 s. The decrease in absorbance at 515 nm was measured after 1, 15, 30, 45, and 60 min. of mixing while using a Photolab 6600 UV-VIS photometer (Xylem Analytics Germany Sales GmbH & Co. KG, WTW, Weilheim, Germany) with toluene as the blank. The scavenging activity of the samples was calculated while using the equation *I* = (*A*<sup>0</sup> − *A*1)/*A*<sup>0</sup> × 100 (%), where *I* is inhibition activity, *A*<sup>0</sup> is the absorbance of the blank and *A*<sup>1</sup> is the absorbance of the mixture.

#### 2.2.4. Interfacial Tension

The interfacial tension of oil-water systems was determined by the pendant drop technique while using an Attension Theta optical tensiometer (Biolin Scientific, Gothenburg, Sweden). The images of the droplets were recorded with a black and white digital camera and the surface tension was obtained by iterative fitting of the shape of the droplet with the Young-Laplace Equation. The droplet was formed while using a 0.718 mm (22 gauge) stainless steel needle.

#### *2.3. Preparation of Emulsions*

The o/w emulsions were formulated with 1, 2, 5, 7.5, 10, or 12 wt % CAS. The CAS solutions were prepared by dispersing the powder in deionized water under gentle stirring at room temperature for 4 h. The sample was left to stand at 4 ◦C overnight to allow for complete hydration and, prior to emulsification, the solution was allowed to equilibrate to room temperature under stirring. Emulsifications were carried out by mixing each of the above CAS solutions and oil phase (5, 10, 20, and 30 wt %; TA or BC oil) at 25 ◦C. The pH of CAS solutions was not adjusted prior to emulsification.

Two different methods were carried out to prepare the emulsions: a) a high-shear homogenization with an rotor-stator device Ultra-Turrax T25 (IKA, Staufen, Germany) for 12 min. at 13, 400 rpm and b) an ultrasonic homogenization (UP400S, Hielscher Ultrasonics, Teltow, Germany) with 400 W, 24 KHz, for 1 min., operated with 100% amplitude. The total sample volume was of 20 mL. Emulsions were kept in the ice bath during sonication. After emulsification, the pH of prepared emulsions was measured while using pH meter (CPH 51, Elteca, Turnov, Czech Republic) and values of 6.44 ± 0.03 and 6.45 ± 0.04 were recorded for TA and BC oil, respectively.

#### *2.4. Characterization of Emulsions*

#### 2.4.1. Size of Emulsion Droplets and Zeta Potential

The size and distribution of emulsion droplets were measured while using the laser diffraction instrument Mastersizer 3000 (Malvern Instruments, Malvern, UK) that was capable of measuring within the size range of 0.01 to 3500 μm. The emulsions were suspended in recirculating milliQ-water flowing through the Hydro SM measuring cell of the instrument, at a pump velocity of 2, 400 rpm. The volume moment mean diameter *D*(4,3) was determined by instrument software according to equation *<sup>D</sup>*(4,3) <sup>=</sup> <sup>Σ</sup>*ni*·*d*4/<sup>Σ</sup> *ni*·*d*3, where *ni* is the number of droplets and *d* represents the droplet diameter. The value of *D*(4,3) corresponds to a mean diameter of spheres with the same volume as the analysed droplets. The emulsions were also observed with optical microscope Zeiss AxioCam MR 5 (Carl Zeiss MicroImaging GmbH, Oberkochen, Germany). Prior to observation, a droplet of emulsion (10 μL) was placed onto a glass microscope slide and then viewed under 10–100× magnification.

The zeta potential of the samples was measured while using a Zetasizer Nano (Malvern Instruments, Malvern, UK) and calculated using the Smoluchowski model. For measurements, 5 μL of the sample were diluted with 1 mL of twice filtered (Millipore, 0.22 μm) deionized water. The average of three records on the freshly prepared samples is reported.

#### 2.4.2. Creaming Index

The stability of emulsions was assessed by visual observation, daily during the first week of preparation and then at seven-day intervals. The emulsions were stored at ambient temperature. The stability was expressed as the height of emulsion layer (*Hemu*l), relative to the total height of the emulsion (*Htotal*), which is referred to as the creaming index (*CI*); (*CI* = *Hemul*/*Htotal* × 100 (%)).

#### *2.5. Antimicrobial Activity*

The antimicrobial activities of oils and emulsions were evaluated while using eight bacterial strains that were obtained from the Czech Collection of Microorganisms (CCM, Brno, Czech Republic). The bacteria were selected to represent major food-borne classes. The gram positive (G+; *Micrococcus luteus* CCM 732, *Staphylococcus aureus* subps. *aureus* CCM 3953, *Bacillus cereus* CCM 2010, *Enterococcus faecalis* CCM 2665) and gram negative (G- ; *Escherichia coli* CCM 3954, *Pseudomonas aeruginosa* CCM 3955, and *Salmonella enterica* subsp. *enterica ser. Enteritidis* CCM 4420, *Serratia marcescens* subsp. *marcescence* CCM 303) strains were employed in the test. All of the microorganisms were maintained on nutrient agar and sub-cultured onto fresh media every two weeks. Suspensions of bacterial strains were prepared by inoculation from pure culture on a Petri plate into sterile tube with nutrient broth and incubation at 30 ◦C for 24 h (*Pseudomonas aeruginosa* and *Bacillus cereus*). Other bacterial strains were incubated at 37 ◦C for 24 h.

The antimicrobial activities of oils and emulsions were determined while using the disc diffusion assay. An overnight culture of each of the bacterial strains in nutrient broth was adjusted to 10<sup>6</sup> CFU/mL and 100 μL of inoculum was spread onto Mueller-Hinton Agar (MHA) sterile agar plates (Hi-Media Laboratories, Mumbai, India). Sterile paper discs (5 mm in diameter) were soaked with 7 μL of oil or with 5 μL of emulsion and then placed on the inoculated agar. As references, discs that were soaked with sterile deionized water and 7.5 wt % aqueous CAS dispersions were used. All of the inoculated plates were incubated at 37 ◦C for 24 h for mesophilic bacterial strains and at 30 ◦C for 24 h for *Pseudomonas aeruginosa* and *Bacillus cereus.* After incubation, the antimicrobial activity of the tested samples was evaluated by measuring the diameter of inhibition zone (mm). The diameter of the disc was subtracted from the diameter of inhibition zone.

#### *2.6. Statistical Analysis*

All the analyses were conducted at least in triplicates, with the Dean-Dixon method being utilized to calculate the means and standard deviations. The Student T-test was applied to detect any statistical differences between the samples (Statistica, StatSoft, Inc., Palo Alto, CA, USA). The P (probability) value of ≤0.05 was considered to be statistically significant.

#### **3. Results and Discussion**

#### *3.1. Properties and Antioxidant Activity of Oils*

#### 3.1.1. Basic Characteristics and Fatty Acid Composition

To evaluate differences between the oils, their properties were determined by measuring their basic characteristics, namely iodine (IV), saponification (SV), acid (AV), and peroxide (PV) values, and via determining their fatty acid composition (Tables 1 and 2). The SV and PV were similar for both oils and they are in agreement with the previously published values. The AVs of TA oil were significantly higher (43.6 <sup>±</sup> 0.6 mg KOH·g<sup>−</sup>1) in comparison with those of BC oil (8.4 <sup>±</sup> 0.7 mg KOH·g<sup>−</sup>1) and they roughly corresponded to the values that were reported in literature [29,30]. In this respect, it can be also stressed that the acid value of TA oil was notably higher than that of the common vegetable oils, thus giving evidence of an increased amount of free fatty acids in this oil. Both of the oils showed high SV, which is a measure of the average molecular mass of fatty acid in the oil. The IV values indicated a higher degree of unsaturation for cumin oil, which was also confirmed by GC analyse, showing a higher content of linoleic acid C18:2 in the sample (Table 1).


**Table 1.** Basic characteristics of tamanu (TA) and black cumin (BC) oils determined as iodine (IV), saponification (SV), acid (AV), and peroxide (PV) values.

The composition of fatty acids in TA oil proved the presence of oleic acid C18:1 (41 wt %) as the dominating fatty acid, followed by linoleic acid C18:2 (28 wt %); stearic C18:0 (16 wt %) and palmitic C16:0 (14 wt %) acids were the major saturates. The minor amounts of C16:1, C17:0, and C20:0 were also detected. In BC oil, linoleic acid C18:2 (57 wt %) was the prevailing fatty acid, followed by oleic C18:1 (25 wt %), palmitic C16:0 (13 wt %), and stearic C18:0 (3.4 wt %) acids (Table 2). The results complied well with data regarding fatty acid composition of the oils reported in literature [29,31,32].

**Table 2.** Composition of fatty acids (wt %) in tamanu (TA) and black cumin (BC) oils determined by Gas Chromatography (GC).


#### 3.1.2. Radical Scavenging Effect of Oils

The DPPH free radical scavenging assay is frequently used to estimate the antioxidant capacity of the substances. In current work, the time development of scavenging effects of BC and TA oils on DPPH• was assayed in toluene and illustrates a slightly higher activity of cumin oil than that of TA oil (Figure 1). After 60 min. of incubation with radicals, 96% of DPPH• were quenched by cumin oil, while TA oil was able to inhibit 94% radicals. The oils were already efficient after 1 min. of incubation time, with 89 and 78% activity for TA and black cumin oil, respectively. The current results are somewhat higher than those that were reported by [4], who used the same procedure and found out that after 60 min. of incubation, 60% of radicals were inhibited by BC seed oil. The scavenging action of both oils can be mainly ascribed to (1) the content and composition of unsaponifiables, (2) the diversity in structural characteristics of phenolic antioxidants present, (3) the synergism of the phenolic antioxidants with other bioactive components, and (4) the different kinetic behaviours of potential antioxidants [4].

**Figure 1.** 1,1-diphenyl-2-picrylhydrazyl (DPPH) free radical-scavenging activities and their development in time determined for BC and TA oils after dissolving in toluene.

#### *3.2. Emulsion Properties, Influence of Processing Conditions, and Composition*

#### 3.2.1. Droplet Size and Distribution

The size of emulsion droplets is an important parameter with a crucial effect on emulsion stability. Light diffraction analyses showed that the emulsion droplet size (*D*(4,3)) was influenced by all of the studied variables, especially by the method of preparation and concentration of stabilizing CAS; and, to lesser extent, by the type and content of both oils.

In this study, the emulsions were prepared by two different emulsification procedures, which involved high-shear homogenization (Ultra-Turrax, UT) or sonicator (US), which obviously controlled the size of emulsion droplets. Emulsification with UT led to coarse emulsions with bigger droplets that ranged from 0.3 to 13 μm, with their properties being notably affected by composition, such as o/w ratio and CAS content (Figure 2A,C). On the other side, the homogenization with US yielded fine emulsions, with *D*(4,3) varying from 0.3 to 0.9 μm and 0.4 to 1.5 μm for BC and TA emulsions, respectively; hence, with a sufficiently small size that is a prerequisite for production of stable systems (Figure 2B,D) [33]. Although droplets after sonication were notably smaller than droplets that were treated with Ultra-Turrax, some differences between emulsions containing BC and TA oils were observed. These can be explained by the character, composition, and physical properties of the oils, which affect the emulsifying efficiency and hence the droplet size of emulsions. Here, an important role can play two crucial characteristics, namely viscosity of the oil and interfacial tension at oil-water interface. The viscosities of the used oils are notably different, TA oil is viscous (20–26 mPa·s) [34], whilst the viscosity of BC oil is lower (6.3 mPa·s) [35]. On the other hand, both of the oils have low and rather similar interfacial tensions at the oil-water interface, and values of 8.4 <sup>±</sup> 1.0 mN·m−<sup>1</sup> and 4.3 <sup>±</sup> 0.5 mN·m−<sup>1</sup> were measured for BC and TA oil, respectively. Therefore, the larger droplets of TA emulsions are likely due to the higher viscosity rather than the differences in the interfacial tension between oils. Interestingly, Wooster et al. also referred that high-viscosity oils formed emulsions with larger droplets than oils with low viscosity [36].

**Figure 2.** Influence of casein sodium salt (CAS) concentration and o/w ratio on volume weighed diameter of emulsion droplets (*D*(3,4)) of (**A**) TA oil emulsions prepared by Ultra-Turrax (UT) (**B**) TA oil emulsions prepared by sonication (US) (**C**) BC oil emulsions prepared by UT, and (**D**) BC oil emulsion prepared by US.

In addition, the droplet distributions in US emulsions were narrower when compared to those that were prepared with UT (Figure 3). This finding is in agreement with studies that were performed on emulsions stabilized by other types of biopolymers, such as CAS/polysaccharides [17,37], whey protein concentrate [38], or on emulsions containing modified starch and maltodextrin [39].

**Figure 3.** Influence of homogenization method on distribution curves of CAS stabilized emulsions. Ultra-Turraxed (UT) and sonicated (US) emulsions are compared across oil type, o/w ratio and CAS concentrations: (**A**) BC emulsions with 2 wt % of CAS (10/90); (**B**) TA emulsions with 5 wt % of CAS (30/70); (**C**) BC emulsions with 7.5 wt % of CAS (20/80); and, (**D**) BC emulsions with 12 wt % of CAS (20/80).

The influence of the processing method on droplet size must always be considered in combination with a composition of emulsions. Regarding the concentration of CAS, the droplet size of UT-prepared emulsions gradually decreased with increasing protein concentration (1 to 5 wt %); however, *D*(4,3) was significantly reduced as CAS concentrations increased to 7.5–12 wt % (Figure 2A,C). Full coverage of droplets with protein occurs at a CAS content of approximately 3 wt % (emulsions with 35 % tetradecane oil phase) [40]. At a higher protein concentration (4 wt %), a higher amount of CAS than required for saturation coverage of the oil droplets is presented; hence, a range of CAS concentrations used in current work is sufficient for completely covering the arising droplets. Therefore, the bigger droplets that were prepared by UT (in the case of sufficient CAS available in the system) can be assigned to the discrepancy between the higher rate of coalescence of oil droplets being formed by a given energy input and the lower rate of CAS adsorption at the oil-water interface during homogenization. Under these conditions, the uncovered droplets tend to come together and form larger droplets again. On the other hand, when the CAS content in aqueous phase is insufficient, flocculation might occur due to casein bridging between droplets, also resulting in rapid creaming [17,20,41,42]. The results from particle sizing were also supported by the distribution curves of UT-emulsions, where a shift in droplet diameter toward smaller sizes was observed as the protein concentration increased, and distribution curves changed from multimodal (1 wt % CAS) to monomodal (10 wt % CAS), which indicated the formation of a stable emulsions (Figure 4A) [43].

**Figure 4.** (**A**) Effect of CAS concentration on distribution curves of BC emulsions with o/w ratio 10/90 prepared by UT and US; (**B**) effect of o/w ratio on distribution curves of BC emulsions with 5 wt % CAS. For tamanu oil, the trend is similar (data not shown).

On the contrary, emulsions that were prepared by US contained much smaller droplets, and only minor changes in the droplet size alongside change in CAS concentration were observed (Figure 2B,D). For emulsions that were prepared with the aid of US, the distribution curves were mostly monomodal with the simultaneous shift in direction to lower droplet sizes (Figure 4A). Figure 5 visualizes an example of light microscopy figure captured on the emulsion prepared with US.

The volume fraction of oil is another parameter with an impact on emulsion droplet size. In the current study, the increasing oil content influenced (*D*(4,3)); however, only at lower concentrations of CAS (1–7.5 wt %) and mainly in the case of UT-prepared emulsions. Here, the droplet size increased with an increasing oil fraction. The effect is documented in Figure 4B, showing the broad distribution curve with three droplet populations being recorded for UT emulsions containing 30 wt % BC oil and 5 wt % CAS; on the other hand, by lowering the oil content to 5 wt %, monomodal distributions were obtained. CAS contents of 10 and 12 wt % afforded droplets that were notably smaller and influence of oil fraction was only marginal. In contrast, the situation was different when it came to emulsions that were prepared by sonication (Figure 4B) and only minor changes in *D*(4,3) and distributions with the increasing oil content were observed at the entire concentration range of CAS used. In systems containing high concentrations of both protein and oil, the emulsions can be stabilized through the formation of a weak particle-based gel network, which slowly reorganizes under the influence of gravity and internal osmotic stresses [40].

Alongside with other variables, the type of oil plays also a role with respect to the droplet size in emulsions, and the droplets of the BC emulsions that were prepared by UT were smaller as compared to droplets containing the TA oil. As already discussed above, the differences can be explained by composition and physical properties of the oils, mainly viscosity and interfacial tension. This difference was again wiped off in emulsions that were prepared using US. The fact that TA oil also contains, in addition to neutral lipids and glycolipids, phospholipids (1.6%), small amounts of sterols, and monoand diacylglycerols, which all have emulsifying properties, can also contribute to easier formation of smaller droplets [9].

**Figure 5.** Optical microscopy (Magnification 20x) of BC emulsions with o/w 20/80 prepared with 1 wt % CAS while using sonication. Scale bar is 10 μm.

#### 3.2.2. Zeta Potential

Zeta (ξ) potential is commonly considered as a stability indicating parameter of dispersion systems. However, the values of ξ potential alone are not always capable of predicting the stability of emulsions, mainly those that are sterically stabilized. Immediately after preparation, the ξ potential ranged from −53 to −41 mV for emulsions with BC oil (prepared with US) and from –60 to –46 mV for those that were prepared with UT. TA emulsions behaved similarly with the potential values lying from –55 to −41 mV and –59 to −44 mV for US and UT emulsions, respectively. In general, UT treatment seemed to deliver slightly lower ξ potentials and, obviously, no or minor systematic influence of oil content on the potential was observed. However, CAS concentrations above 5 wt % might have a positive influence in potential lowering, mainly for emulsions with TA oil (Figure 6). The negative charges of emulsion droplets are, therefore, due to the presence of CAS stabilizer. As such, this protein is negatively charged at a natural pH and its aqueous dispersion in the absence of oil showed a potential of −18.7 mV.

In current study, ξ potential measurements were conducted on emulsions with non-adjusted pH (TA oil emulsions pH = 6.44 ± 0.03, BC oil emulsions pH = 6.45 ± 0.04). At a pH close to caseinate isoelectric point (pH 4.6), the CAS emulsions are unstable [44], thanks to a reduction in electrostatic repulsion between the droplets. However, at pH 6, the negative ξ potential of CAS assured emulsion stability due to electrostatic and steric stabilization mechanisms.

**Figure 6.** Zeta potential of freshly prepared emulsions at native pH of ~6,5 as a function of o/w ratio and CAS content (**A**) BC emulsions prepared with Ultra-Turrax (UT), (**B**) BC emulsions prepared with sonication (US), (**C**) TA emulsions prepared with UT, and (**D**) TA emulsions with US.

Differences in the ξ potentials between emulsions that were prepared by UT and US can be assigned to combined effect of sonication treatment and oil properties. When compared with many other proteins, caseins are particularly disordered and substantially hydrophobic, which assist in their rapid adsorption to oil droplet surfaces during emulsification. In aqueous solutions, CAS forms a random coil with only small amount of secondary structure due to the relatively high content of hydrophobic amino acids (proline residues) [45]. During sonication, the hydrophobicity of CAS changes, thanks to the different orientation of proteins at the oil-water interface, which can impact the ξ potential of emulsions formed by US treatment [46]. The impact of oil type on ξ potential, which is observed in our work, was also reported for soy bean protein isolate emulsions containing medium chain triacylglycerols, palm, soybean, and rapeseed oils after US treatment [47]. The lowering of ξ potential was ascribed to the formation of protein aggregates and different amounts of protein adsorbed onto emulsions containing studied oils of different compositions.

#### 3.2.3. Phase Studies and Emulsion Stability

When proteins are used to stabilize emulsions, the main mechanism governing their creaming in the presence of non-adsorbed stabilizer is related to depletion flocculation [48]. Kinetic stability of emulsions, as described by creaming index (CI), provides indirect information on droplet flocculation and destabilization occurring in an emulsion.

Not all of the emulsions prepared in the current study demonstrated kinetic stability, even immediately after preparation. When observed by naked eye, some of them formed a bottom serum layer and a small cream phase on the top. On the other hand samples with specific formulations, mainly with high CAS content, exhibited good resistance to creaming. Figure 7 shows the CI of the freshly prepared emulsions. Stable emulsions (CI = 100%) were observed at CAS concentrations of 10–12 wt % (UT emulsion) and 7.5–12 wt % CAS (US emulsions). However, samples containing lower concentrations of CAS destabilised readily and the extent of destabilization, expressed as CI, was obviously dependent on oil content. More specifically, the CI increased with an increasing amount of oil, which was more notable on UT emulsions in comparison with emulsions that were prepared by US. Therefore, the formation of stable emulsions was considered to be a synergic effect of the US processing, resulting in small droplets with narrower distribution and the sufficient amount of CAS. Only a minor difference in CI was observed between emulsions with TA or BC oils.

The long-term stability of the emulsions was evaluated by the recording of changes in their visual appearance during storage until the point of emulsion breaking appeared. Not surprisingly, the most stable emulsions were those that contained the highest protein content (12 wt % CAS). For example, BC emulsions with 30 wt % oil prepared by both procedures remained unchanged for seven days of storage with CI being of 100%. Stability studies also corroborated the fact that US treatment assured the production of emulsions with limited creaming. At a CAS concentration of 10 wt %, the emulsions that were manufactured with US were stable towards creaming for seven days, whilst in emulsions that were prepared with UT, the creaming already started three days after preparation. Moreover, UT manufactured emulsions with even lower CAS content were prone to breakdown and underwent phase separation approximately after a week of storage. Interestingly, in most emulsions, although the CI value was lower immediately after emulsification, it did not notably change during their storage. From the CI data, it was also apparent that the creaming process took about a week to be completed, thereafter compaction of the cream layer started, which stopped when no more oil droplets and proteins could be packed in the top cream layer. The compaction/compression of the cream layer was also reported by [49]. These results can be generalized for both BC and TA emulsions. The above observations demonstrate that an important role in stability of CAS-containing emulsions is related to concentration of this biopolymer in the aqueous phase, which closely correlates with emulsion viscosity. The viscosity of CAS emulsions is controlled by the interactions between the droplets and mainly by the nature and strength of interparticle attractive forces, which depends on the structure of the CAS layer that was adsorbed at the oil-water interface. Additionally, the CAS self-assembly and aggregation of non-adsorbed CAS in the aqueous phase play an important role with respect to the viscosity and stability behaviour of CAS stabilized emulsions [50].

The data from the observation of creaming were in reasonably good correlation with particle sizing measurements, which were also conducted during emulsion storage. The results revealed that the *D*(4,3) increased during a month-long storage at an ambient temperature. The US emulsions changed their droplet sizes to higher extent than the UT systems; however, emulsions that were prepared with the aid of UT were more prone to phase separation and their breakdown occurred earlier. Here again, the key role played the content of CAS in emulsions. Emulsions with 1 or 2 wt % CAS were stable in terms of size only during the first days, emulsion with higher CAS content for a week and emulsions with 10 and 12 wt % CAS remained unchanged for a longer period of time (Figure 8). This is in agreement with findings in [51]. However, when considering the practical application of TA and BC emulsions, their long-term stability is not satisfactorily. In this respect, the prolongation of sonication time or two-stage homogenization with an additional treatment of coarse emulsions with high-pressure homogenizer can be used for the improvement of stability. The effect of high-pressure homogenization on properties of CAS emulsions studied authors [33]. They compared properties of thus prepared emulsions with emulsions that were prepared by Ultra-Turrax and concluded that UT produced larger droplets and less stable systems than the homogenizer did. However, during homogenization, the choice of correct pressure was essential, as the over-processing of emulsions and conformation changes of proteins can occur at higher pressures [52].

**Figure 7.** Comparison of creaming index (CI; CI—100% is for stable emulsion) determined on freshly prepared emulsions, as affected by processing method and composition of emulsions (**A**) BC emulsions prepared with Ultra-Turrax (UT), (**B**) BC emulsions prepared with sonication (US), (**C**) TA emulsions prepared with UT, and (**D**) TA emulsions with US.

**Figure 8.** Long term stability of TA and BC emulsions stabilized with CAS determined at room temperature: development of volume weighed diameter of emulsion droplets (*D*(4,3)) in time.

#### *3.3. Antimicrobial Activity*

Illnesses that are caused by pathogenic bacteria and/or their toxins are of great concern to public health. For the suppression of microorganism growth, plant extracts with antimicrobial activity can also be used [4].

Antibacterial activity of oils and emulsions (o/w 30/70) prepared thereof with 2 and 7.5 wt % CAS was determined while using the disc diffusion method and expressed in terms of the size of the inhibition and halo zones (mm).

Disc diffusion assay proved the inhibitory activity of both oils and selected emulsions against the gram positive bacteria, which was controlled by the type of the oil and the formulation of the emulsion. Regrettably, none of the samples were capable of suppressing the growth of the tested gram-negative species (*Escherichia coli, Pseudomonas aeruginosa, Salmonella enterica,* and *Serratia marcescens*). Table 3 shows that the effect of TA oil was significantly higher (p ≥ 0.5) than that of BC oil, both for *B. cereus* and *S. aureus*. It can be seen from the bigger inhibition zones with the sizes ranging from 8.2 ± 0.4 mm (*B. cereus*) to 6.0 ± 1.2 mm (*S. aureus*). This oil also caused lower bacterial growth (14.8 ± 1.3 to 7.7 ± 0.7 mm) around the inhibition zones (halo zone) and it was able to act, at least partially, against *M. luteus* and *E. faecalis* (halo zone only). The effects of TA oil encapsulated in emulsions with 2 and 7.5 wt % of CAS were rather similar, although the activity of emulsion stabilized with 2 wt % CAS was comparable with that of non-encapsulated TA oil (p ≥ 0.5). The antibacterial activity of the BC oil was, in comparison with TA oil, weaker, and this oil only showed antibacterial activity against *B. cereus* (5.7 ± 1.5 mm) and *S. aureus* (4.2 ± 0.7 mm). Emulsion that was stabilized with 2 wt % CAS containing BC oil exerted the comparable effect, which was weaker than that of BC oil alone. Neither BC oil nor its emulsions were efficient against *M. luteus* and *E. faecalis*. The example of growth media images with inhibition zones observed after application of TA and BC emulsions are shown in Figure 9. The lower activity of oils and emulsions against gram negative bacteria is a result of composition in their cell walls, as they contain an outer lipopolysaccharide membrane, which more efficiently protects the bacteria from the disruption caused by oils. As reported in literature, the antimicrobial activity of BC oil mainly results from the presence of thymoquinone, p-cymene, longifolene, and thymohydroquinone. Moreover, unsaturated, long chain fatty acids, such as linoleic and oleic acids present in the oil, were also reported to possess antibacterial and antifungal activity [8]. The study

conducted by [4] supported the above results regarding the bioactivity of BC oil, also extending the range of its inhibitory activity to action against yeasts, namely *Saccharomyces cerevisiae* and *Candida albicans*. Regarding triacylglycerol-based TA oils, their bioactivity was thoroughly investigated by [1], who reported that TA oil exhibited two distinct antibacterial effects: the first against gram positive bacteria acting by direct inhibition of mitotic growth and the second potent effect against gram negative strains due to an increased release of β-defensin 2 peptide by macrophages. However, the results of our study have not confirmed the effect against gram negative strains.

**Table 3.** Antibacterial activity of BC and TA oils and their 30/70 o/w emulsions against Gram positive strains; No inhibition zone towards *Micrococcus luteus* (*M. luteus*) and *Enterococcus faecalis* (*E. faecalis*) were observed.


*n.d.* not determined.

**Figure 9.** Images of growth media with inhibition zones observed after application of TA and BC emulsions: from "1" clockwise: TA oil with 2 wt % CAS, BC oil with 2 wt % CAS, TA oil with 7.5 wt % CAS, BC oil with 7.5 wt % CAS; in the middle is negative control.

#### **4. Conclusions**

In this work, sodium caseinate was used to stabilize emulsions containing bioactive tamanu and black cumin oils. The emulsions were prepared by ultrasound treatment or high-shear homogenization with Ultra-Turrax. The analysis of fatty acid composition in the oils revealed a higher degree of unsaturation for cumin oil with higher content of linoleic acid C18:2, which corresponds to the higher iodine value determined for this oil. Both of the oils effectively scavenged DPPH radicals, thus showing antioxidant activity. The results revealed that, under emulsification, sonication was the more efficient procedure and it afforded emulsions with a small particle size (0.3 to 1.5 μm) throughout the entire used concentration ranges of oils and caseinate. In addition to the emulsification technique, the ability to form stable emulsions of small, initial droplet sizes was mainly controlled by concentration of stabilizing caseinate and, to lesser extent, by the type and amount of used oils. In comparison with sonicated emulsions, the properties of emulsions that were prepared with Ultra-Turrax depended, to a higher extent, on their composition and the emulsions were more prone to destabilization. The oils and their selected emulsions exhibited antibacterial activity against gram positive strains (*S. aureus* and *B. cereus*); regrettably, the gram negative species were fully resistant against their action. Of the studied oils, the tamanu oil and its emulsions were more efficient. The antibacterial and antioxidant properties of the oils, together with their beneficial fatty acid composition, make them suitable for use as carriers for various lipophilic bioactive substances; the incorporation in caseinate emulsions further increases their applicability in hydrophilic systems.

**Author Contributions:** Conceptualization, L.U., V.K. and P.E.; methodology, L.U., V.K., P.E., O.R. and E.K.; validation, L.U. and V.K.; formal analysis, L.U., O.R. and E.K.; investigation, L.U. and V.K.; resources, L.U. and V.K.; writing—original draft preparation, L.U.; writing—review and editing, L.U. and V.K.; visualization, L.U. and V.K.; supervision, V.K.

**Funding:** This work was supported by the Czech Science Foundation (19-16861S) and by the Ministry of Education, Youth and Sports of the Czech Republic – Program NPU I (LO1504). One of us (LU) also appreciates support of the internal grants of TBU in Zlín IGA/CPS/2019/004 funded from the resources of specific academic research.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Communication* **In-Vitro Hemocompatibility of Polyaniline Functionalized by Bioactive Molecules**

#### **Kateˇrina Skopalová 1, Zdenka Capáková 1, Patrycja Bober 2, Jana Pelková 3,4, Jaroslav Stejskal 2, V ˇera Kašpárková 1,5, Marián Lehocký 1,5, Ita Junkar 6, Miran Mozetiˇc <sup>6</sup> and Petr Humpolícek ˇ 1,5,\***


Received: 6 September 2019; Accepted: 1 November 2019; Published: 11 November 2019

**Abstract:** Hemocompatibility is an essential prerequisite for the application of materials in the field of biomedicine and biosensing. In addition, mixed ionic and electronic conductivity of conducting polymers is an advantageous property for these applications. Heparin-like materials containing sulfate, sulfamic, and carboxylic groups may have an anticoagulation effect. Therefore, sodium dodecylbenzenesulfonate, 2-aminoethane-1-sulfonic acid and *N*-(2-acetamido)-2-aminoethanesulfonic acid were used for modification of the representative of conducting polymers, polyaniline, and the resulting products were studied in the context of interactions with human blood. The anticoagulation activity was then correlated to surface energy and conductivity of the materials. Results show that anticoagulation activity is highly affected by the presence of suitable functional groups originating from the used heparin-like substances, and by the properties of polyaniline polymer itself.

**Keywords:** polymer conductivity; conducting polymer; polyaniline; hemocompatibility

#### **1. Introduction**

The hemocompatibility of any material is a crucial factor in its application in biomedicine, regenerative medicine, or biosensors. The phenomenon of hemocompatibility is very complex and comprises various processes of which blood coagulation remains at the center of attention when tissue engineering and biosensing applications are considered. Any material that comes in contact with blood should not induce coagulation either by disrupting physiological blood flow or by interactions of the blood with the surface of a material on a molecular level.

In some special applications (e.g., biosensing or tissue engineering of electrosensitive tissues) the conductivity of materials can be an advantage. However, common conducting materials (such as metals or metal oxides) can induce an undesired reaction on contact with tissues (e.g., because of their different elasticity compared to native tissue). Therefore, conducting polymers can be more appropriate. Moreover, conducting polymers have a huge potential for the mentioned applications as they show mixed ionic and electronic conductivity, which is preferred for electrically active biointerfaces. The effect of conductivity on the biological systems is a comprehensive issue. We, therefore, refer to the publication of Rivnay et al. [1], who discuss the topic of ionic and electronic conductivity in depth.

Polypyrrole (PPy), poly(3,4-ethylenedioxythiophene) (PEDOT) and polyaniline (PANI) can be considered as the most frequently studied conducting polymers. The biocompatibility of these polymers was recently studied in many research works. For example, the Ramanaviciene et al. [2] tested the effect of PPy particles on mouse peritoneum cells. The action of injected PPy particles was dependent on their concentration and duration of treatment. Nevertheless, the amount of neutrophils did not exceed the limit values at any of the particle concentrations used. Vaitkuviene et al. [3] also investigated the effect of PPy nanoparticle concentration on living systems. It has been demonstrated that PPy nanoparticles exhibit a cytotoxic effect on murine embryonic stem cells, murine hepatoma cells, and human Jurkat T lymphocytes at concentrations higher than 19.4 <sup>μ</sup>g·mL<sup>−</sup>1. From a practical point of view, the polymer films are more important as regards bio applications. Vaitkuviene et al. [4] coated gold-plated glass slides by PPy. Proliferation of mouse stem cells on the treated surfaces was comparable to reference (tissue plastic). The PANI, which is the object of this work, was also previously studied in the context of biocompatibility. Its biocompatibility and especially the cytotoxicity were tested on pristine PANI salt and base [5], as well as on its globular and nanotubular forms [6]. In the scientific literature, it is often mentioned that PPy exhibits favorable properties compared to PANI. However, the study of Humpolícek et al. [ ˇ 7] dealing with the comparison of the biocompatibility of PANI and PPy in terms of cytotoxicity and embryotoxicity showed that the form of polymer (salt vs. base) is more important than its type (PPy vs. PANI). Likewise, the present study compares, inter alia, two forms of PANI, salt and base.

The amounts of information found in the scientific literature about the hemocompatibility of conducting polymer differ. Whilst PPy has been intensely studied, predominantly in functionalized forms or as a part of composites, PEDOT or polyaniline has been investigated as possible biomaterials but their hemocompatibility has not been studied. Li et al. [8] dealt with the blood compatibility of PPy films functionalized by heparin. The compatibility was evaluated using plasma re-calcification time and platelet adhesion. The results showed that, with immobilized heparin, platelet adhesion and platelet activation on PPy films were significantly suppressed and, moreover, the plasma re-calcification time was considerably prolonged.

Ferraz et al. [9] tested the hemocompatibility of nanocellulose/PPy membranes modified by a stable heparin coating. The results demonstrated that the heparinized composites were comparable with hemocompatible polysulfone regarding platelet adhesion and thrombin generation, whereas, in terms of complement activation, they were more biocompatible than commercially available membranes. Mao et al. [10] investigated *O*-butyryl chitosan-grafted PPy film for blood compatibility using platelet-rich plasma. The surface with immobilized *O*-butyryl chitosan exhibited lower platelet adhesion and fibrinogen adsorption compared to unmodified PPy. The films showed good blood compatibility and high electrical conductivity.

Previously it was mentioned that the functionalization of neat PANI films with poly(2-acrylamido-2-methyl-1-propanesulfonic acid) (PAMPSA) results in the hindering of blood coagulation [11]. Humpolícek et al. [ ˇ 11] followed up the studies of Paneva et al. [12] and Yancheva et al. [13] who have shown that PAMPSA can act against blood clotting in a similar way as heparin, either alone or incorporated in copolymers. The biological activity of PAMPSA was studied by Šorm et al. [14] who found out that some methacrylic copolymers that contain (similarly to heparin) sulfate, sulfamic, and carboxylic groups have an anticoagulation effect. Therefore, the compounds holding some of these functional groups were chosen for the functionalization of neat PANI films in this work. Dodecylbenzenesulfonic acid sodium salt (SDBS), 2-aminoethane-1-sulfonic acid (taurine), and *N*-(2-acetamido)-2-aminoethanesulfonic acid (ACES) were, therefore, chosen as modifiers with a potential anticoagulation effect.

#### **2. Materials and Methods**

#### *2.1. Sample Preparation*

PANI films containing the above-given compounds (Figure 1) were prepared using two different procedures: (1) By modification of the surfaces of neat PANI films, PANI salt, or PANI base with SDBS, taurine or ACES, and (2) by adding the respective substance directly into a reaction mixture of aniline hydrochloride and ammonium peroxydisulfate used for the preparation of PANI.

**Figure 1.** The formula of (**A**) PANI hydrochloride, (**B**) SDBS, (**C**), taurine, and (**D**) ACES.

Neat PANI films were prepared according to a standard procedure described by Stejskal and Gilbert [15]. Aniline hydrochloride (2.59 g, Penta, Czech Republic) was dissolved in water to a 50 mL solution, and ammonium peroxydisulfate (5.71 g, Lach-Ner, Czech Republic) was similarly dissolved to a 50 mL solution. Both solutions were mixed and immediately poured onto the polypropylene foil and into polyethylene terephthalate (PET) blood-collection tubes. After 1 h, the reaction mixture was removed and the resulting films of green conducting salt (PANI-S) deposited on the foils/walls were rinsed with 0.2 M hydrochloric acid, followed by methanol, and were left to dry in air. Some of the films were deprotonated by immersion in 1 M ammonium hydroxide for 12 h and thus converted to blue, non-conducting films of PANI base (PANI-B). In order to prepare reprotonated films, the samples of neat PANI-B were exposed to 2 wt % solutions of SDBS, taurine or ACES (all from Sigma-Aldrich, St. Louis, MO, USA). Please note that the reprotonation of PANI-B with the individual compounds manifests itself by the reverse change in color from blue to green. The process was slow and took several weeks as a rule (for reprotonation with ACES it took three months). Then, the solutions were removed, and the films were rinsed with methanol and left to dry in air. Modified PANI-S films were prepared by simple pouring each of the 2% solutions of SDBS, taurine or ACES onto the neat film. After 24 h, the solutions of bioactive substances were removed, and the films rinsed with methanol and dried in air. The samples were labelled as PANI-BSDBS, PANI-BTaurine, PANI-BACES, PANI-SSDBS, PANI-STaurine, and PANI-SACES.

The second type of PANI film was prepared by adding the respective substance into a mixture of aniline hydrochloride and ammonium peroxydisulfate. Aniline hydrochloride (2.59 g) was again dissolved in 50 mL of an aqueous solution of SDBS or taurine (40 g·L<sup>−</sup>1). Then, the aqueous solution (50 mL) of ammonium peroxydisulfate (5.71 g) was added, and the reaction mixture was stirred and poured over the substrates (polypropylene foil and blood-collection tubes). The reaction was left to continue for 1 h. The films of PANI salts formed on the substrates containing respective dopants were then rinsed with 0.2 M HCl and were left to dry in the air. The samples were labelled as PANI-MSDBS and PANI-MTaurine. Polyaniline powders obtained after polymerization along with the films were treated similarly and used for conductivity determination.

#### *2.2. Surface Energy Measurement*

Contact angle data was obtained with a Surface Energy Evaluation System (Advex Instruments, Brno, Czech Republic). Deionized water, ethylene glycol, and diiodomethane were utilized as test liquids. The volume of droplets was set to 5 μL for all experiments to avoid errors connected with gravity acting on the sessile drop. Five contact angle readings were averaged to obtain one representative value. The free energy of the film surfaces was determined by the Lifshitz–van der Waals "acid–base" model. Total surface energy (γtot) was calculated and reported.

#### *2.3. Conductivity*

The conductivity of the polyaniline powders collected after polymerization and compressed to pellets was measured by the four-point van der Pauw method. A programmable electrometer with an SMU Keithley 237 current source and a Multimeter Keithley 2010 voltmeter (Keithley instrument, Cleveland, OH, USA) with a 2000 SCAN 10-channel scanner card were employed. Measurements were carried out at ambient temperature.

#### *2.4. Anticoagulation Test*

In all tests, venous blood was collected from healthy donors by venipuncture using the vacuum blood collection system into the 5 mL collecting tubes (VACUETTE, Greiner Bio-One, Kremsmünster, Austria) after obtaining informed consent. All tests were conducted in accordance with the Helsinki Declaration. Plasma was prepared from the venous blood by centrifugation (15 min, 3000× *g*). Plasma was subsequently transferred to the blood-collection tubes coated with studied PANI films. The measurement of coagulation was performed using an instrument commercially used in hospitals, the SYSMEX CA-1500 (Siemens, Erlangen, Germany). The principle of measurement is based on the change of turbidity (measured at a wavelength of 660 nm) as a consequence of adding coagulation reagent which induces the formation of fibrin clothes. In the present study, the following coagulation parameters in human blood plasma treated with 0.109 M citric were studied: (1) Thrombin clotting time (TCT), (2) activated partial thromboplastin time (aPTT), and (3) prothrombin time (PT).

#### **3. Results and Discussion**

Hemocompatibility is a complex process and its testing by common in-vitro methods is not able to comprehensively cover all its aspects. Blood coagulation as a whole includes not only the coagulation cascade but also the platelet adhesion and can be, for example, triggered by various biochemical cues or blood rheology (flow velocity, flow turbulence, shear stress) [16]. Especially in the case of platelet adhesion, the shear rates of the fluid are important. In this work, we focused on one of the important aspects of blood coagulation, namely on the detection of coagulation induced by the surface properties of modified PANI films. Although the biological properties of PANI in its native or modified form have already been studied, and cytotoxicity [5], interaction with stem cells [17], or interaction with tissues have already been reported, another important parameter of biocompatibility, namely interaction with blood has only been the subject of a few studies. The effect of PANI itself on platelet adhesion, hemolysis and plasma recalcification time has been studied by Li et al. [18]. Here, PANI-coated polyurethane (PU) fibers were tested. Platelets on PANI-coated PU demonstrated lower aggregation (6.87 <sup>×</sup> 105 cm−2) than platelets on PU fibers without modification (15.63 <sup>×</sup> <sup>10</sup><sup>5</sup> cm<sup>−</sup>2). As regards hemolysis, according to requirements of ISO 10993-4, materials with hemolysis values <5% are considered safe in contact

with blood. In the study, PANI-PU fibers exhibited hemolysis values of 0.14% and non-coated PU fibers 0.21%. The tests also showed that PANI prolonged plasma recalcification time by 13 s, improving thus anticoagulation. The impact of PANI on the blood coagulation can be also improved by modification of its surface by bioactive substances. For example, a work of Zhang et al. [19] studied modifications of PANI films with poly(ethylene glycol) (PEG) to reduce the adhesion of proteins and platelets on its surface. In this case, the PEG chain was covalently bound to PANI film. Adhesion of bovine serum albumin (BSA) and γ-globulin decreased with increasing concentration of surface-bound PEG. The same results were obtained for platelet adhesion. Li and Ruckenstein [20] tested BSA and platelet adhesions on native PANI and poly(ethylene oxide) (PEO) grafted films. Correspondingly to the previous study, modification of the PANI film with hydrophilic PEO also reduced protein adhesion, in this case by 80%. Platelet adhesion also decreased significantly. In the case of the unmodified film, there was adhesion of 18 <sup>×</sup> <sup>10</sup><sup>3</sup> mm<sup>−</sup>2. For the modified film, the adhesion value was 2.1 <sup>×</sup> <sup>10</sup><sup>3</sup> mm<sup>−</sup>2. Similar results were achieved in work of Humpolicek et al. [11]. This report investigated the impact of PANI functionalized with PAMPSA on blood coagulation and found out that PANI films reprotonated with this high-molecular-weight acid hindered blood coagulation by interaction with three of the coagulation factors: Xa, Va, and IIa. Moreover, PAMPSA-modified PANI also reduced platelet adhesion. It was generally expected that methacrylic copolymers containing similarly to heparin the sulfate, sulfamic, and carboxylic groups, may have an anticoagulation effect. With respect to this assumption, SDBS, taurine, and ACES might have the potential to influence blood compatibility as well. As PANI is a versatile material that can easily be modified by a variety of methods, the PANI films were prepared using the above-mentioned substances either by their direct addition into the reaction mixture with subsequent preparation of films, or by modification of neat PANI-S and PANI-B films via pouring over the solutions of each of these substances.

The standard polyaniline salt, PANI-S, contains chloride counter-ions (Figure 1A), as well as sulfate ions originating from the reduction of peroxydisulfate. These counter-ions are removed after the conversion to PANI-B. After immersion of PANI-B into the solution of, e.g., taurine, immobilization of this substance onto PANI film takes place resulting in their mutual salt (PANI-BTaurine) containing corresponding taurine counter-ions. As regards PANI-S films modified with studied heparin-like substances, it can be assumed that a layer of the respective substance remains on the surface of neat PANI. When PANI is prepared in the presence of taurine in reaction mixture, PANI-MTaurine, the films contain chloride, sulfate, and taurine counter-ions. The PANI films prepared in various ways thus differ in the type and occurrence of counter-ions, or more general in surface chemistry.

Blood coagulation can be triggered by contact with interfaces due to the following three crucial characteristics/properties of a surface: (1) The surface energy, (2) the acidity (pH), and (3) the interaction of surfaces with coagulation factors of blood. The adherence of plasma proteins and the subsequent adherence of platelets and leukocytes are also of importance regarding the plasmatic coagulation, which again depends on the surface characteristics of the material. The surface energies measured on modified PANI films were similar for almost all samples (Table 1) and did not differ from γtot determined on neat PANI-S and PANI-B films [21]. The exception here was PANI-BACES with γtot equal to 41 <sup>±</sup> 1 mN cm<sup>−</sup>1. For practical application, the stability of the samples is crucial, and the effect of aging on the surface energy was also determined. The surface energy of all samples was stable for at least two weeks, except for the PANI-BTaurine and PANI-STaurine films (Table 1).

It is known that the physiological pH of blood lies within a tight range from 7.31 to 7.42 [22], and several non-physiological reactions can occur as a consequence of the pH changes, including inhibition and prolongation of coagulation processes. The pH of blood after being in contact with the samples was therefore determined, and it was unambiguously confirmed that the pH was not notably influenced by any of the tested samples.


**Table 1.** The surface energy of modified PANI surfaces, γtot, determined immediately after preparation, and after 7 and 14 days on the stored under laboratory conditions. The values are compared with γtot of neat PANI-S and PANI-B films.

(a) Reproduced from [21]; n.d. not determined.

Conductivity is another important characteristic of conducting polymers. The standard films of PANI-S commonly show conductivity within units of S·cm−<sup>1</sup> [23]. The conductivity measurements on fresh samples, and after 7 and 14 days of their storage at room temperature are summarized in Table 2. A certain conductivity drift is likely to be associated with still decreasing sample humidity. Measurements revealed that only the samples synthesized with SDBS or taurine present in the reaction mixture showed reasonably high conductivity and all other films exhibited conductivity lower than 10−<sup>4</sup> S·cm−1. It is beyond the scope of this manuscript to define the relationship between the conductivity of materials and their applicability in biomedicine. This is a comprehensive issue and the situation is complicated not only by the fact that conducting polymers combine ionic and electronic conductivity but also by the different resistivity of various tissues (e.g., from 100 Ω·cm for blood to <sup>1000</sup> <sup>Ω</sup>·cm for some tissues) [24] which corresponds to the conductivity of 0.01 to 0.001 S·cm<sup>−</sup>1. In this context, the conductivity of here-tested PANI samples, especially these where taurine was present in the reaction mixture, is significantly higher reaching ~18 S·cm<sup>−</sup>1.

**Table 2.** The conductivity of bulk PANI (S cm<sup>−</sup>1) containing the studied substances.


The coagulation parameters of the blood after contact with the studied surfaces are presented in Table 3. The thrombin clotting time (TCT), activated partial thromboplastin time (aPTT), and prothrombin time (PT) were studied as clinically relevant parameters using the methods described earlier [11]. Similarly to neat PANI-S and PANI-B [11], none of the PANI modified with SDBS, taurine or ACES significantly impacted blood coagulation (Table 3), and all the coagulation parameters were within their respective physiological ranges. The expected impact of taurine, SDBS, or ACES on the hemocompatibility was not hence confirmed. So far, the only compound suitable for influencing the hemocompatibility of PANI is, as previously described [11], PAMPSA. Here, an interesting question arises with regard to the reason why SDBS, taurine, or ACES were not able to modify the hemocompatibility of PANI even though they contain functional groups similar to PAMPSA.


**Table 3.** Impact of PANI with modified surfaces on selected coagulation parameters, prothrombin time (PT), activated partial thromboplastin time (aPPT), thrombin clotting time (TCT), expressed as times to the coagulation start.

Note: As reference coagulation parameters of donor blood were used. Normal ranges for coagulation parameters of a healthy person: PT 11.0–13.5, aPTT 25–32, TCT below 20 s. The values are expressed as mean value ± standard deviation of three tests.

Based on the results presented in Tables 1–3 and on previously published findings, the reason for this can be found in the low-molecular-weight character of these substances, which in all three cases lies below 350 g·mol−1. On the other hand, in the case of hemocompatible PANI-PAMPSA, its anticoagulation activity can be also assigned to the presence of the high-molecular-weight PAMPSA polyanion (molar mass <sup>≈</sup> 106 g·mol−1) bound to polycationic PANI surface. During reprotonation, a fraction of anions on a PAMPSA chain is bound as counter-ions to the PANI backbone. However, a substantial fraction of the functional sulfo groups still remains free to interact with blood and can act as an anticoagulant agent. In this context, it can be mentioned that also the best-known anticoagulation agent, heparin, is a polymer. This polysaccharide consists of chains containing 1,4-bonded residues of uronic acid and D-glucosamine, and its molar mass ranges from 3000 to 30,000 g·mol−1, with an average of 15,000 g·mol−<sup>1</sup> [25]. However, its anticoagulant activity is closely related to a unique pentasaccharide sequence necessary for linking to antithrombin [26]. In the heparin chain without this pentasaccharide sequence, the anticoagulation activity is absent [27].

In summary, the present pilot study covers only a limited part of the complex hemocompatibility process. The impact of above-mentioned parameters, such as blood flow velocity, flow turbulence, shear stress, or tension on the hemocompatibility of PANI-based surfaces has not been studied so far. It, therefore, opens the possibility for further testing of the effect of the low-molecular-weight substances with sulfo groups on the hemocompatibility, which can be conducted under physiological flow conditions and using blood flow models.

#### **4. Conclusions**

The PANI films modified by substances with anticipated anticoagulant activity, sodium dodecylbenzenesulfonate (SDBS), 2-aminoethane-1-sulfonic acid (taurine) and *N*-(2-acetamido)- 2-aminoethanesulfonic acid (ACES) have been studied. The hemocompatibility tests conducted on these PANI films confirmed that none of them showed anticipated anticoagulation activity, though the functional groups typical for anticoagulation substances were present. The absence of the activity can be ascribed to the low molecular weight of these compounds used for PANI modification. It can be, therefore, concluded that, in addition to the presence of suitable functional groups in the molecule, the dopants introducing anticoagulation activity should exhibit polymer-like character with sufficiently high molecular weight.

**Author Contributions:** Conceptualization, P.H., M.L., V.K., J.S., M.M. and P.B.; methodology, Z.C., K.S., I.J.; formal analysis, Z.C.; investigation, J.P. and K.S.; resources, P.H. and M.L.

**Funding:** This work was funded by the Czech Science Foundation (19-16861S) and the Ministry of Education, Youth and Sports of the Czech Republic (NPU I, LO1504). One of us, K.S., acknowledges the support of an internal

grant from TBU in Zlín (IGA/CPS/2019/004) financed from funds of specific academic research. Authors M.L., I.J., and M.M. would like to express their gratitude to the Slovenian Research Agency (P2-0082).

**Acknowledgments:** Authors thank Lada Utˇekalová and Petra Rejmontová for technical support.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Preparation of Progressive Antibacterial LDPE Surface via Active Biomolecule Deposition Approach**

**Salma Habib 1, Marian Lehocky 2,3, Daniela Vesela 2, Petr Humpolícek ˇ 2,3, Igor Krupa <sup>1</sup> and Anton Popelka 1,\***


Received: 9 September 2019; Accepted: 15 October 2019; Published: 17 October 2019

**Abstract:** The use of polymers in all aspects of daily life is increasing considerably, so there is high demand for polymers with specific properties. Polymers with antibacterial properties are highly needed in the food and medical industries. Low-density polyethylene (LDPE) is widely used in various industries, especially in food packaging, because it has suitable mechanical and safety properties. Nevertheless, the hydrophobicity of its surface makes it vulnerable to microbial attack and culturing. To enhance antimicrobial activity, a progressive surface modification of LDPE using the antimicrobial agent grafting process was applied. LDPE was first exposed to nonthermal radio-frequency (RF) plasma treatment to activate its surface. This led to the creation of reactive species on the LDPE surface, resulting in the ability to graft antibacterial agents, such as ascorbic acid (ASA), commonly known as vitamin C. ASA is a well-known antioxidant that is used as a food preservative, is essential to biological systems, and is found to be reactive against a number of microorganisms and bacteria. The antimicrobial effect of grafted LDPE with ASA was tested against two strong kinds of bacteria, namely, *Staphylococcus aureus* (*S. aureus*) and *Escherichia coli* (*E. coli*), with positive results. Surface analyses were performed thoroughly using contact angle measurements and peel tests to measure the wettability or surface free energy and adhesion properties after each modification step. Scanning electron microscopy (SEM) and atomic force microscopy (AFM) were used to analyze the surface morphology or topography changes of LDPE caused by plasma treatment and ASA grafting. Surface chemistry was studied by measuring the functional groups and elements introduced to the surface after plasma treatment and ASA grafting, using Fourier transform infrared (FTIR) spectroscopy and X-ray photoelectron spectroscopy (XPS). These results showed wettability, adhesion, and roughness changes in the LDPE surface after plasma treatment, as well as after ASA grafting. This is a positive indicator of the ability of ASA to be grafted onto polymeric materials using plasma pretreatment, resulting in enhanced antibacterial activity.

**Keywords:** biointerface; polyethylene; plasma treatment; antibacterial; grafting modification

#### **1. Introduction**

Low-density polyethylene (LDPE) is the most common industrial polymer and is mainly used in food packaging because of its useful properties, such as its ease of shaping, handling, and recycling, and its high cost efficiency [1–4]. Nevertheless, LDPE lacks proficiency in traits such as printability, adhesion, and some other surface properties, as it has an inert surface with a very low surface free energy (wettability). Changing the bulk properties of LDPE by mixing additives is not recommended for food packaging applications; therefore, surface modification is a safe, easy, and cost-effective option [5,6]. Many studies have enhanced the surface properties of LDPE using physical or chemical methods together with conventional methods such as flame treatment, chemical grafting, irradiation, free-radical mechanism, and corona and radio-frequency (RF) plasma treatments [7–11]. These plasma treatments have been found to be more effective techniques for the modification of the surface properties of polymers without any disruption of bulk polymer properties [12,13]. Plasma, as the fourth state of matter, is able to initiate the ionization of air species and surface components, converting them into electrons and negative, positive, and neutral ions, in addition to metastable and free radicals [13–15]. The free radicals that are generated are readily reactive, with mainly oxygen-containing functional groups [16]. This results in enhancement of the wettability, adhesion, roughness, and reactivity properties. The plasma treatment methods vary from corona (atmospheric pressure) to vacuum-based RF plasma. All types were found to have similar effectiveness on the enhancement of the surface properties of polymers [17], even when the ionized gas species varied [18]. For nonthermal, low-temperature, and cold plasma, no heat is generated or required, and they only affect a layer of a few tens nanometers depth on the surface. Therefore, these methods are widely used in medical applications and medicine in general [16,19–21]. Plasma is also known to be safe for the environment and for human health, as it kills microorganisms and cleans medical equipment, with no significant impact on human cells [21–23].

In food packaging, microbial and bacterial fouling is a critical concern, and a large amount of research has been devoted to the use of antibacterial packaging materials to prevent the vulnerability and susceptibility of food to any type of microorganism and to increase the shelf life of the food [1,24,25]. Some antibacterial techniques involve mixing of the antibacterial agent within the polymeric material to generate biomaterials [26,27]. However, this process is not suitable for all packaging materials because it changes the main functional and mechanical properties of the materials and decreases their stability. Other studies have applied the antibacterial agents by surface grafting or surface tangling, with attachment on the surface achieved by chemical grafting, coating, or plasma treatment [28,29]. To enhance the antibacterial properties, different chemicals or nature-based compounds, such as chitosan [30–33] and alkyl pyridiniums [34], were grafted onto polymer surfaces and proved to reduce the number of, or entirely kill, bacteria (both gram-positive and gram-negative species). Other chemicals known as organic acids, which are mostly used as preservatives, were tested for their antibacterial activation when attached to polymers. Polyacrylic acid, for instance, was grafted with chitosan on LDPE and tested against the strong bacteria *Escherichia coli* (*E. coli*) [15]. These acids showed successful results in decreasing and eliminating the presence of bacteria on the LDPE surface.

Ascorbic acid (ASA), commonly known as vitamin C, is an organic acid known to be an antioxidant. It possesses two hydroxyl groups that can be deprotonated and is an effective radical scavenger [35,36]. It is an essential vitamin needed to maintain body health [37]. There are different studies on the antimicrobial effect of ascorbic acid that provide positive results. It was found to inhibit bacterial growth and prevent biological infections [38–40]. Further investigations found antiviral [40,41] as well as antifungal activities [40], alone or in combination with additional agents, to exert a synergetic effect. Tests with *E. coli*, *Staphylococcus aureus* (*S. aureus*), and some other types of bacteria [42–45] demonstrated the very effective ability of vitamin C to inhibit the growth of, and kill, bacterial colonies by penetrating bacterial walls, affecting the metabolism with no harmful effect on human cells [43]. ASA has the ability to enter a cell and modify its redox reaction through its hydroxyl groups, which eventually leads to the inhibition of microorganism growth; thus, ASA can be considered a good antimicrobial agent [46,47].

In this study, ASA was used for the preparation of a progressive LDPE surface through plasma-assisted grafting, which has excellent antimicrobial properties. The antibacterial effectiveness of the LDPE surface modified by ASA was tested against *E. coli* and *S. aureus*.

#### **2. Materials and Methods**

#### *2.1. Materials*

Commercial grade low-density polyethylene (LDPE) FE8000 was supplied in pellet form by Qatar Petrochemical Company (QAPCO, Doha, Qatar). Thin homogeneous films approximately 0.4 mm thick were prepared by compression molding using an industrial mounting press machine (Carver, Wabash, IN, USA). The pellets were melted at 160 ◦C and compressed for 2 min using a force of 2 tons, while maintaining the set temperature to obtain a film with the desired smooth surface. The samples were then cooled to room temperature by water. The LDPE films were cleaned by acetone to remove any additives, residuals, or any possible contaminations from the molding process that might affect the surface properties, and were then dried in an air atmosphere for 20 min at room temperature. Small strips (5 cm × 1 cm) were cut out and directly used for the surface treatment and subsequent analyses.

Ethylene glycol (>98% FLUKA, Morris Plains, NJ, USA), formamide (>98% FLUKA, Merelbeke, Belgium), ultra-pure water (prepared by Purification System Direct Q3, Millipore Corporation, Molsheim, France), and acetone (99.9% Scharlau, Barcelona, Spain) were used as testing liquids for wettability analyses.

L-ascorbic acid (>99.0% Research-Lab, Uran Islampur, India) molecular weight = 176.14 g/mol was used as an antimicrobial agent.

#### *2.2. Plasma Treatment of LDPE*

Plasma treatment of LDPE films was performed using a Venus 75-HF enclosed low-temperature plasma-generating system (Plasma Etch Inc., Carson, CA, USA). Plasma-excited species were generated using a radio-frequency (RF) generator operating at a frequency of 13.56 MHz. The chamber of the plasma system was evacuated to a pressure level of approximately 0.2 Torr using a rotary vacuum pump before plasma ignition. Optimization of the treatment process was carried out by varying the nominal power, treatment time, and working gas to obtain the maximum level of hydrophilicity on the LDPE surface. The applied nominal power varied from 50 W to 120 W, and the treatment time ranged from 10 s to 180 s at a constant optimal nominal power of 80 W. The gas flow rate was 10 cm3/min. The film surfaces were treated from both sides in air.

#### *2.3. Antibacterial Agent Grafting*

Immediately after the plasma treatment, the LDPE samples were immersed in a 10 vol % aqueous solution of ASA. The immersion process was continuous for 24 hours at 24 ◦C to achieve radical grafting. ASA is converted to an ascorbate radical by electron donation to a radical [48], namely, the alkoxyl radical present in the plasma-treated LDPE surface created by the decomposition of hydroperoxide. The ascorbate radicals can then interact with the double bonds present in plasma-treated LDPE created by disproportionation reactions; therefore, ASA can be covalently grafted onto the LDPE surface (Figure 1). After the grafting process, the LDPE samples were thoroughly washed with water and ethanol to remove weakly bound or unreacted ASA from the LDPE surface.

**Figure 1.** Scheme of ascorbic acid (ASA) grafting on low-density polyethylene (LDPE) via plasma treatment.

#### *2.4. Hydroperoxide Determination*

Iodometric titration was performed to determine the concentration of all hydroperoxide species accumulated on the surface of LDPE after plasma treatment. Plasma-treated LDPE samples were placed into a covered Erlenmeyer flask, which was filled with 50 mL of glacial acetic acid. An excess (1.0 g) of sodium iodide was added, and the flask was purged with argon gas for 15 min to eliminate interactions with air. After 15 min, the well-stirred mixture became yellow (oxidation of iodide to iodine by hydroperoxides incorporated on the LDPE surface) and was titrated with a 0.0005 M sodium thiosulfate pentahydrate aqueous solution. The reactions were carried out in an argon atmosphere and protected from light. The hydroperoxide concentration on the LDPE surface was calculated per area considering two treated sides of the LDPE samples. The titration was repeated 3 times to obtain average values and to ensure reliable results.

#### *2.5. Surface Wettability Measurements*

The changes in hydrophilicity induced by plasma treatment of LDPE films were evaluated by static contact angle measurements using the sessile drop method. An OCA35 surface free energy analysis system (DataPhysics, Filderstadt, Germany) equipped with a CCD camera was employed for this purpose. Water, formamide, and ethylene glycol were used as testing liquids to evaluate the total surface free energy and polar and dispersive components using the conventional Owens–Wendt–Rabel–Kaelble method. A droplet of approximately 3 μL of each testing liquid was placed on the air-facing samples. The contact angle was calculated after approximately 3 s to allow thermodynamic equilibrium between the liquid and the sample interface to be reached. The reported value for each testing liquid corresponds to the mean of at least five measurements taken on different parts of the substrate surface.

#### *2.6. Graft Yield Analysis*

Graft yield measurements were used to prove the grafting of ASA on the LDPE surface. The graft yield of modified LDPE was calculated by gravimetric measurements. The graft yield (GY) was calculated by Equation (1):

$$GY[\%] = ((\mathcal{W}\_2 - \mathcal{W}\_1) / \mathcal{W}\_1) \cdot 100\% \tag{1}$$

where *W*<sup>1</sup> and *W*<sup>2</sup> represent the weights of the LDPE samples before and after the modification.

#### *2.7. Film Thickness Investigation*

Thickness measurements were carried out by an F20-UVX film thickness analyzer (Filmetrics, San Diego, CA, USA) to analyze the thickness of plasma-affected and ASA-modified layers of the LDPE surface. The film thickness value was evaluated based on the differences in reflectance (%) between reference and measured samples in wavelength range of 190–1700 nm. LDPE substrate (4.5 mm

thick, with refractive index of 1.5) and LDPE substrate (4.5 mm thick, with refractive index of 1.4, considering polar functional groups) were used as reference samples for the thickness measurements of plasma-treated and ASA-modified LDPE layers. Analysis of plasma-treated and ASA-modified LDPE samples was performed in air atmosphere. The spectrum was analyzed by varying the measured parameters to obtain the best fit between the theoretical and measured data using FILMeasure software, v7.19.0. Readings from five different areas were captured for each sample, and a mean value was evaluated.

#### *2.8. Peel Test*

A 90◦ peel test was performed to measure the adhesion characteristics of LDPE samples in terms of the peel resistance using a Lloyd 1K Lf plus-UTM standard testing machine (Lloyd Instruments, West Sussex, UK). Samples 19 mm in width and 6 cm in length were attached on a polypropylene tape containing poly(2-ethylhexyl acrylate) adhesive (Scotch tape). The test was undertaken with Scotch tape pressed on top of the treated LDPE surface. The unbonded end of the testing tape was peeled off at 90◦ at a crosshead speed of 10 mm/min. The test was stopped after 6 min when the tape was complexly detached from the LDPE surface, and 6 separate readings were carried out to obtain average values of the peeling force.

#### *2.9. Surface Chemistry Characterization*

Fourier transform infrared spectroscopy with attenuated total reflectance (FTIR-ATR) was used to qualitatively investigate the chemical composition changes of plasma-treated LDPE surfaces. An FTIR Spectrometer Frontier (PerkinElmer, Waltham, MA, USA) equipped with a ZnSe crystal was used for these analyses to capture data from a penetration depth of 1.66 μm. Spectra in the wavenumber range of 4000–550 cm−<sup>1</sup> were obtained using an average of 8 scans, with a resolution of 4 cm<sup>−</sup>1.

The chemical composition changes caused by corona treatment of the LDPE surface were quantified by X-ray photoelectron spectroscopy (XPS). An AXIS XPS system (Kratos Analytical, Manchester, UK) was used for this study. The XPS system contains a spherical mirror analyzer and a delay-line detector for fast screening of the chemical composition, ensuring high spectral resolution and sensitivity. This system allows the analysis of data at a sampling depth of 1–10 nm.

#### *2.10. Surface Morphology Analysis*

The surface morphology of LDPE samples was analyzed by scanning electron microscopy (SEM). This technique allowed us to obtain information about surface morphology changes after each modification step. For this purpose, a Nova Nano SEM 450 microscope (FEI, Hillsboro, OR, USA) was employed. A thin Au layer a few nanometers thick was sputter-coated on the LDPE samples to obtain high-resolution images with high magnification (20,000×) and to avoid the accumulation of electrons on the measured layer.

Detailed information about the three-dimensional changes in the surface topography of the LDPE samples was obtained using atomic force microscopy (AFM). An MFP-3D AFM device (Asylum Research, Abingdon, Oxford, UK) was employed in these experiments. Scanning was carried out under ambient conditions by a silicon probe (Al reflex-coated Veeco model, OLTESPA, Olympus, Tokyo, Japan) in tapping mode in air (AC mode), allowing images with a surface area of 1 <sup>×</sup> 1 <sup>μ</sup>m2 to be obtained. Moreover, the roughness parameter value (Ra) was calculated from AFM images obtained from Z-sensor.

#### *2.11. Antibacterial Tests*

A modified ISO 22196, an internationally recognized test method was used to evaluate the antibacterial activity of modified plastic materials (and other nonporous surfaces of products) to inhibit the growth of, or kill, test microorganisms [49]. The LDPE samples were first disinfected by UV radiation and then placed in sterile Petri dishes. This was followed by inoculation of the samples (25 <sup>×</sup> 25 mm2) using 0.1 ml of standardized bacteria suspension of *S. aureus* (CCM 4516, 1.8 <sup>×</sup> 106 cfu/mL) and *E. coli* (CCM 4517, 1.4 <sup>×</sup> 107 cfu/mL). The samples were covered by disinfected polypropylene foil (20 <sup>×</sup> 20 mm2) with 70% ethanol. Incubation of the inoculated samples was performed at 95% of relative humidity at 35 ◦C for 24 hours. The polypropylene foil was then removed, and LDPE samples were imprinted on plate count agar (3 times on different areas) and incubated at 35 ◦C for 24 hours. Then, the results were read, and the increase in the number of bacterial colonies was evaluated based on scaling from 0 to 5, where 0 represents the best antimicrobial effect, with no growth of bacteria colonies. An additional incubation at 35 ◦C for 24 hours was followed by final reading and evaluation of the results. All of these analyses were performed using 3 different LDPE samples to ensure reliable antimicrobial efficiency results.

#### **3. Results**

#### *3.1. Hydroperoxide Concentration*

Plasma treatment was used as an effective tool for generation of active species in the LDPE surface necessary for the subsequent grafting process by ASA. As the plasma treatment introduces polar functional groups onto the surface by radicalization, different kinds of functional groups can be found (mainly oxygen-containing groups). Through exposure to air during and after plasma, most of the free radicals are converted into peroxides [50]. However, it is difficult to distinguish the amount of peroxide functionalities in either the infrared (IR) spectroscopy or XPS O1s shift spectra; thus, a classic quantification method by iodometric titration according to Wagner and Thelen [51,52] was used to obtain valid concentration values. In Figure 2, LDPE samples were treated with air plasma at different exposure times. By applying iodometric titration, it was found that the hydroperoxide concentration increased as the treatment time increased from 10 s (7.6 <sup>×</sup> 10−<sup>8</sup> mol/cm2) to 60 s (9.0 <sup>×</sup> 10−<sup>8</sup> mol/cm2) of exposure to air plasma. An additional increase in treatment time did not lead to another increase in hydroperoxide concentration. This proves that exposing the polymer samples for a longer time does not increase the formation of hydroperoxides; thus, at the optimum time, the surface would be saturated with a certain amount of peroxides [50]. From this observation, an optimum time for achieving the maximum hydroperoxide concentration was evaluated, and LDPE samples were treated with plasma at 60 s prior to the ASA grafting process.

**Figure 2.** The peroxide concentration of plasma-treated LDPE samples.

#### *3.2. Surface Wettability Analysis*

The changes in the surface wettability of the modified samples were analyzed through contact angle measurements, which are shown in Figures 3 and 4 and summarized in Tables 1 and 2. Surface free energy and wettability are indicators of the ability of the liquid surface to be attached to the solid surface. This indicates that the lower the contact angle of a sample is, the higher its wettability is. To evaluate surface free energy and its components, water (surface free tension = 72.1 mN/m, polar component = 52.2 mN/m and dispersive component = 19.9 mN/m) [53], ethylene glycol (surface free tension = 48.0 mN/m, polar component = 19.0 mN/m, and dispersive component = 29.0 mN/m) [54] and formamide (surface free tension = 56.9 mN/m, polar component = 33.4 mN/m, and dispersive component = 23.5 mN/m) [53] were used. Untreated LDPE has hydrophobic properties and is chemically inert; thus, its wettability is low under basic conditions. Its water contact angel was 95.7◦, with a low total surface free energy (29.3 mJ/m2) and insignificant polar component (1.9 mJ/m2). These results refer to the hydrocarbon skeleton –C–H, which has poor reactivity, and thus, no polarity was observed.

The initiation of reactions on the surface by RF plasma improved the polarity by inducing the surface through radicalization. The introduction of new oxygen-containing functional groups helped to increase the total surface free energy to 49.0 mJ/m2, and the contact angle of water decreased to almost half, with a value of 50.0◦. The wettability increased as a result of the new polar functional groups on the surface. Plasma treatment affected the LDPE surface only at very small depth (28.2 nm), which was confirmed by film thickness measurements. Grafting of the antibacterial agent enhanced the polarity even further because of its effective side and defined structure attachment. The ASA-grafted LDPE surface using plasma treatment exhibited the lowest value for the contact angle of water (32.3◦); therefore, the highest value for the total surface free energy (67.0 m J/m2) and its polar component (63 mJ/m2) was achieved. The introduced functional groups on the LDPE surface were able to effectively react and form new bonds with ASA, and the graft yield was 0.4%, indicating a multilayer formation of ASA. This was confirmed also by film thickness measurements, where the film thickness was 10.1 nm. The graft yield and film thickness analyses confirmed a formation of ASA multilayered structures on the LDPE surface. The effect of plasma treatment on the grafting of ASA onto the LDPE surface was shown, in comparison with untreated LDPE subjected to modification by ASA with subsequent thorough washing. In this case, the contact angle and the total surface free energy were similar to those of untreated LDPE.


**Table 1.** The contact angles and graft yields of LDPE samples.


**Table 2.** The surface free energy of LDPE samples.

**Figure 3.** The contact angles of testing liquids on LDPE samples.

**Figure 4.** The surface free energy of LDPE samples.

#### *3.3. Adhesion Analysis*

Adhesion properties depend on the wettability and surface morphology (roughness) of a material surface. Roughness occurs as a result of physicochemical interactivity or chemical composition on the surface. Adhesion properties can be effectively analyzed by measurements of the peeling resistance. Figure 5 shows the changes in peeling resistance and the changes in adhesion of LDPE after plasma treatment and ASA grafting. Higher resistance induces higher adhesion, which is an outcome of a rougher surface and better wettability. The untreated LDPE surface exhibited relatively poor adhesion, and therefore, the peel resistance reached a value of 40.5 N/m because of the smooth surface and low wettability. The plasma-treated samples showed significant enhancement in the adhesion of the LDPE surface. The peel resistance increased significantly to 83.5 N/m after plasma treatment. This increase in peel resistance was affected mainly by the increase in the wettability and surface roughness caused by the incorporation of polar functional groups and etching reactions, respectively. ASA grafting onto the LDPE surface led to even higher peel resistance (97.3 N/m) because the highest wettability was achieved. ASA was also subjected to LDPE modification with and without application of plasma

treatment to study the effect of plasma treatment on the covalent grafting of ASA on the LDPE surface. The untreated LDPE with ASA showed similar peel resistance (45.2 N/m) to the untreated LDPE sample, indicating the lack of ASA after thorough washing.

**Figure 5.** Peel resistance of LDPE samples.

#### *3.4. Chemical Composition Investigation*

The chemical composition of the LDPE samples after each modification step was analyzed using Fourier transform infrared (FTIR) spectroscopy. The FTIR spectra of untreated, plasma-treated, and modified LDPE samples are shown in Figure 6. The FTIR spectrum of untreated LDPE is characterized by specific absorbance bands attributed only to nonpolar hydrocarbons in the main chain and branches. These bands represent carbon–carbon and carbon–hydrogen vibrations. Thus, –C–H stretching vibrational bands are observed at 2915 cm−<sup>1</sup> and 2847 cm−1, and bending and scissoring vibrations are observed at 1473 cm−<sup>1</sup> and 717 cm−1, respectively. Plasma treatment of LDPE was responsible for the introduction of new functional groups by radicalizing some carbons on the polyethylene surface. This process allows the interaction between the plasma-activated surface and air elements. Oxygen-containing groups are the main functional groups following these interactions. However, these new functional groups were not clearly detected by FTIR spectroscopy because of the relatively high penetration depth of the IR beam when using a ZnSe crystal (1.66 μm) compared with the thickness of the plasma-affected layer, which was only a few tens of nm. The incorporation of new oxygen-containing groups was clearly confirmed by XPS analyses. ASA is an organic compound with ether, carbonyl, and 4-hydroxyl groups apparent in the oxygen-containing spectral region. New vibrational absorbance bands appeared in the FTIR spectrum of LDPE grafted by ASA, where –OH was represented by a broad absorbance band between 3650 cm−<sup>1</sup> and 3150 cm<sup>−</sup>1, C=O was observed at 1623 cm<sup>−</sup>1, –COOH was observed at 1776 cm<sup>−</sup>1, and C–O–C was observed at 1060 cm−1. This could indicate the presence of ASA on the LDPE surface after the modification process.

For the quantification of the chemical composition of LDPE samples, the XPS technique was employed. The XPS spectra of LDPE samples after each modification step are shown in Figure 7. The XPS spectrum of untreated LDPE consists mainly of the C1s peak, with 98 at.% at a binding energy of ~285 eV. It also contains low-intensity O1s (2.6 at.%) and N1s (0.2 at.%) peaks associated with oxygenand nitrogen-containing functional groups, originating from processing additives or intermediates coming from air interactions with the LDPE surface. Moreover, other peaks were observed in XPS spectra, which are associated with auger electrons, and therefore they were disregarded from total

atomic %. Plasma treatment of the LDPE surface was responsible for the incorporation of new functional groups, as indicated by a significant increase in the intensity of the O1s peak at a binding energy of ~530 eV, achieving a value of 12.4 at.%. In addition, the intensity of the N1s peak increased in the XPS spectrum of LDPE after plasma treatment because of the incorporation of some nitrogen-containing groups (C–N or C–NH3 <sup>+</sup>). This led to a reduction in the intensity of the C1s peak to 86.5 at.% due to the removal of some carbons during etching, radicalization, and replacement with oxygen-containing groups. ASA grafting on the LDPE surface led to an increase in the at.% of the O1s peak, which was attributed mainly to the oxygen-containing groups in ASA. The ASA-grafted LDPE samples showed an O1s peak with 15.8 at.%, whereas the intensity of the C1s peak decreased to 83.1 at.%. Furthermore, the N1s intensity remained unchanged in comparison with that of the plasma-treated LDPE sample (1.1%), as ASA does not contain any nitrogen functional groups.

**Figure 6.** Fourier transform infrared spectroscopy (FTIR) spectra of LDPE samples.

**Figure 7.** X-ray photoelectron spectroscopy (XPS) spectra of LDPE samples.

#### *3.5. Surface Morphology Analysis*

The changes in the surface morphology and topography of the LDPE samples after each modification step were studied through SEM and AFM, respectively. SEM and AFM images of LDPE samples are shown in Figure 8. The untreated LDPE surface exhibits characteristic texture and morphology originating from the production process. Plasma treatment did not cause any significant changes in the surface morphology obtained by SEM from larger surface areas. On the other hand, the LDPE samples grafted with ASA experienced clear changes in their surface morphology, showing bulges and valleys in the functionalized regions on the surface as grafting took place.

**Figure 8.** Scanning electron microscopy (SEM), 3D height. and amplitude atomic force microscopy (AFM) images with line profiles (Z-sensor) of LDPE: (**A**) untreated; (**B**) plasma-treated; (**C**) ASA-grafted. Note: Ra represents the roughness parameter.

AFM was used to determine the topographical and roughness changes (Ra) in the LDPE surface that occurred after plasma treatment and ASA grafting in the small surface area (1 <sup>×</sup> <sup>1</sup> <sup>μ</sup>m2). Moreover, this technique was used to obtain line profiles, clearly indicating specific nanopattern dimensions. As reported in Figure 8, the Ra value of the untreated LDPE surface was only 2.5 nm. An application of plasma treatment led to an increase in surface roughness, while Ra increased by almost 80% to 4.4 nm. The increase in roughness can be attributed to the etching process during plasma treatment, which led to nanosurface topography changes. Grafting of ASA onto the LDPE surface resulted in a less rough surface (Ra = 2.8 nm) in the small surface area, but with a specific texture belonging to the created ASA layer.

#### *3.6. Antibacterial Analysis*

The antimicrobial activities of the LDPE tested against gram-positive *S. aureus* and gram-negative *E. coli* using intensive microbial activity assays are summarized in Table 3 and are shown in Figure 9. The untreated LDPE showed no resistance or inhibition to bacterial growth. This was because of its poor inhibition properties resulting from the chemical composition of LDPE. Plasma treatment of

LDPE was responsible for the low resistance to bacterial growth on the surface. On the other hand, ASA grafted on the LDPE surface showed a high ability for inhibition of *S. aureus*. ASA proved to be highly active against *S. aureus*, similar to its use alone in free form [19,23,38]. ASA could have affected the protein on the bacterial wall, affecting bacterial growth because of its ability to lower the pH and cause instability of bacterial cell membranes [55]. After ASA was grafted onto the LDPE surface, the inhibition activity against *S. aureus* was successful, with over 80%–90% total inhibition. Interestingly, the inhibition of *E. coli* growth was not as intense, although a clear reduction in colony growth and reproduction was observed, as shown in Figure 9.


**Table 3.** Antimicrobial activity of LDPE samples.

<sup>1</sup> The scale for assessing the growth of bacterial colonies: 0—without growth; 1—detectable amount (single colony); 2—detectable amount (combined colony); 3—second imprint, distinguishable colonies, third imprint can be detected;

4—third imprint, distinguishable colonies; 5—overgrown, continuous growth.

**Figure 9.** Example of total microbial counts of LDPE samples on plate count agar with inoculated bacteria: (**A**) untreated (*S. aureus*); (**B**) ASA-grafted (*S. aureus*); (**C**) untreated (*E. coli*); (**D**) ASA-grafted (*E. coli*).

#### **4. Conclusions**

In this study, ascorbic acid (ASA or vitamin C) was grafted onto an LDPE surface via plasma treatment in order to improve the antimicrobial effect. Plasma treatment was effectively used as a radical initiator with subsequent incorporation of ASA, which served as an antimicrobial agent, on the LDPE surface. This modification was confirmed by enhanced wettability and adhesion properties. The presence of ASA on the LDPE surface after the grafting process was confirmed by chemical composition analyses. Chemical composition and surface morphology or topography analyses were

used to confirm the presence of ASA on the LDPE surface. The significant antimicrobial effect of such modified LDPE against gram-positive *S. aureus* was demonstrated, with an inhibition efficiency of over 80–90%.

**Author Contributions:** Conceptualization, A.P.; methodology, A.P.; validation, A.P.; formal analysis, A.P.; investigation, S.H., D.V., and A.P.; resources, M.L., I.K., and P.H.; data curation, A.P.; writing—original draft, S.H.; writing—review and editing, A.P., M.L., and I.K.; visualization, A.P. and S.H.; supervision, A.P.; project administration, A.P.; funding acquisition, A.P. and I.K.

**Funding:** This publication was made possible by Award JSREP07-022-3-010 and NPRP10-0205-170349 from the Qatar National Research Fund (a member of The Qatar Foundation). The statements made herein are solely the responsibility of the authors.

**Acknowledgments:** The authors gratefully acknowledge Gas Processing Center (GPC), Qatar University, for carrying out the XPS analysis.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
